Notes: Update

SOME MONTHS AGO in this space I made the point that much of the social-science research that gets noticed in the press—not all the research, by any means, but a lot of it—seems to trumpet conclusions that to many people would seem obvious. I went on to express gratitude that at least one branch of academic inquiry wasn’t calling into question everything about life that we take for granted. The gratitude may have been premature.

It is not that in these past many weeks the volume of obvious research findings has diminished. I have a growing stack of newspaper clippings and press releases that summarize a wide variety of new and unsurprising research. “Differences are found in what disturbs men and women.”“Bereaved children reported feeling more scared and less happy than their counterparts.” “Most [marital] problems start after the ‘honeymoon’ period of the marriage is over.” “Children’s reaction to pain not influenced by ethnic background.” “Men reported telling lies [in order to have sex] significantly more frequently than women.” The studies keep getting churned out, and what I imagine to be the average person’s basic view of the way the world works keeps getting confirmed.

No, the problem has been with the mail. For a while, it seemed, every day brought another fat, menacing envelope, its embossed ritual markings (“Department of Psychology,” “Department of Sociology”) heralding another intemperate screed from someone with an advanced degree. The authors argued that a great deal of social-science research comes to conclusions that are not at all obvious; that even when the conclusions do seem obvious, the reason may be “hindsight bias” or even “self-deception”; and that uninformed criticism is of benefit to no one. In support of their contentions they cited an array of research and observed, variously, that misery doesn’t necessarily love company, that opposites don’t necessarily attract, and that familiarity doesn’t necessarily breed contempt. Each letter aspired to a sad, Olympian disdain that was frequently achieved and remarkable to experience. Whenever one arrived, I would immediately put down the newspaper (“Top diet tack: Eat sensibly”) and get the unpleasantness over with.

ONE DAY RECENTLY what appeared to be yet another piece of invective, bearing the usual forty-five cents’ worth of postage and a nervous-making cancellation mark (“Stanford, CA”), arrived in the mail. To my surprise, it turned out to be a friendly letter from a man named N. L. Gage, who is the Margaret Jacks Professor of Education (emeritus) at Stanford University. He allowed that he, too, had long been interested in what he called the “obviousness question” in social science, and he enclosed a copy of a scholarly paper, “The Obviousness of Social and Educational Research Results,” which he was about to publish in a journal. He said he and I had come to “well-nigh opposite conclusions,” but he thought I might enjoy his paper anyway.

Well, I did, at first. Gage begins with a remark from the 1930s attributed to the novelist James T. Farrell, in which a sociologist was defined as “someone who will spend $10,000 to find the location of the nearest house of ill fame.” He quotes James B. Conant, the longtime president of Harvard University, who in the 1950s characterized much social science as consisting of “for the most part highly limited and unsystematized generalizations, which are the stock in trade of everyday life for all sane people.” Gage retails similar criticisms by other estimable thinkers up through the mid-1980s, and admits, by way of summary: “These opinions are hard to ignore.”

But then he asks, in effect: What is meant by “obvious”? Does the charge that a proposition is obvious even have a useful meaning? Gage cites a doctoral dissertation written by a Stanford student, Daphna Baratz, who took sixteen actual social-science propositions—for example, “People who go to church regularly tend to have more children than people who go to church infrequently”; “Single women express more distress over their unmarried status than single men do”—and asked a group of some forty Stanford undergraduates to indicate, using a numbered scale, just how obvious each proposition was. She also told them that each statement was true (though in fact only half of them were). Baratz then gathered a second group of forty Stanford undergraduates, presented them with the opposite of those sixteen propositions, told them that the statements were true, and asked them to rate the statements for obviousness. As you will perhaps have anticipated, the true and false findings were deemed by the two groups of students to be just about equally obvious. This experiment was extended and confirmed bv another Stanford doctoral candidate, Lily Wong, in 1987. “From the work of Baratz and Wong,” Gage writes, “we can conclude that the feeling that a research result is obvious is untrustworthy.”

The tendency of people to regard as obvious any research finding they happen to read about is, Gage admits, only a tendency. And some research findings are certainly both very true and truly obvious. It is disconcerting, though, to have the notion of obviousness, of all things, called into question, and Gage’s proposal for a fresh research effort that will focus on obviousness— who knows what will turn up?—is an additional cause for alarm. Gage himself does not seem to be especially worried about just how far the concept of obviousness will ultimately unravel: “Most adults’ generalizations about human interactions,” he points out in his paper, “are at least functional.” Let us hope that that generalization, too, is functional.

—Cullen Murphy